CSD/ - Contains structural coordinates for a second generation Feringa motor (150441) [@doi:10.1126/science.aam8808] from the Cambridge Structural Database. I can't recall the mapping between the deposited coordinates and the sets listed in the SI. I believe that the set listed in the SI do not correspond to the motor in Figure 2 of the paper, however. For example, the first descriptor for all the motors in the main text begin with S and their quantum scan coordinates all begin with R. These isomers should not interconvert. The coordinates in the CSD visually conform to the upper left panel of Figure 2.
torsionfit/ - Cloned from https://github.com/choderalab/torsionfit to investigate using psi4.
torsion-scan/
scan.in - Example psi4 input file to scan the dihedral in increments of 10 degrees (self-contained) with output in scan.out after a convergence issue. I investigated this issue only lightly, but I did raise the point with the developers of psi4 back in November.generated-from-chaya-torsion-fit/
torsion-drive.ipynb - Notebook containing a few helper functions to setup and start psi4 calculations for fixed scan coordinates (instead of using the built-in psi4 driver method, like in scan.in mentioned above).B3LYP-torsion-scan/ - Coordinates (pdb) and psi4 input files (dat) split into subdirectories by the value of the dihedral.geometry-optimization/
optimize-with-frozen-dihedral.in - A test to see if optimizing the initial structure works correctly (in this case: doesn't crash). The title string says "HF/cc-pVDZ" but the commited code is at just "B3LYP/3-21G".optimize.png - The results of the optimization.gaussian-torsion-scan-HF-631G/ - Gaussian input (gjf) and output (log) files for a basic ground state scan of the Feringa motor, using the coordinates specified inside the input file at HF/6-31G without the checkpoint file. Reported total CPU time of 1 day, 17 hours. The scan energies can be pulled from the output file by using GaussView or grepping for the energies: grep "SCF Done: E(RHF) =" 1540441.log | awk {'print $5'} and the scan coordinates: grep "22,3,5,24" 1540441.log | tail -n +2 | awk {'print $4'} (note that the scan coordinates need to be wrapped). Edit: actually this isn't working at the moment because the log file contains more "SCF Done" entries than I expected, so those are not the final, optimized energies. The energy scan is plotted in gaussian-torsion-scan.ipynb and high-energy.mol2 is one of the high energy conformers.
gaussian-torsion-scan-HF-631Gdp/ - Gaussian input (gjf) and output (log) files for a basic ground state scan of the Feringa motor, using the coordinates specified inside the input file at HF/6-31G(d,p) without the checkpoint file. Reported total CPU time of 5 days, 16 hours. There's a jupyter notebook in here that has the energies plotted.
gaussian-torsion-scan-excited-HF-631Gdp/ - Gaussian input (gjf) and partial output (log) files for an excited state scan of the Feringa motor, using the coordinates specified inside the input file at HF/6-31G(d,p) and using keyword cis(root=1). This job didn't finish after 200 hours and the rwf scratch file was over 326 GB, so it's probably not the proper way to scan the excited state.